issues
search
whitebox-research
/
excursions
1
stars
1
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
What is attribution patching?
#20
kgreyy
opened
1 week ago
1
How can you implement attribution patching?
#19
Zmavli
opened
1 month ago
1
How can one tell if a transformer has learned the closed form of the Fibonacci sequence as opposed to brute memorization?
#18
vorrjjard-2
opened
1 month ago
0
How does tokenization (and variant tokenizations) interact with lexical disambiguation?
#17
Zmavli
opened
1 month ago
1
What (philosophically valid) method could be used to tell whether or not a neuron (or set of neurons/feature/head) is related to lexical disambiguation of some words?
#16
Zmavli
opened
1 month ago
0
How does a transformer perform addition?
#15
vorrjjard-2
opened
1 month ago
0
How can you create a dataset to examine lexical disambiguation?
#14
Zmavli
opened
1 month ago
2
How can you implement ablation without referring to existing tutorials?
#13
Zmavli
opened
1 month ago
0
How do you load `InceptionV1`?
#12
zrkrlc
closed
1 month ago
0
Can you do activation patching on InceptionV1?
#11
zrkrlc
opened
1 month ago
1
Will the transformer overfit on just the Fibonacci sequence?
#7
kgreyy
opened
1 month ago
0
What is the effect of the choice of padding strategy on transformer performance?
#6
kgreyy
opened
1 month ago
1
How do you train an SAE and use it to get a neuron out of superposition?
#5
MostDeadDeveloper
opened
1 month ago
3
How do you get a neuron out of superposition?
#4
MostDeadDeveloper
closed
1 month ago
0
How do you train a sleeper agent model?
#3
MostDeadDeveloper
closed
1 month ago
1
What is the meaning of life, the Universe, and everything else?
#1
zrkrlc
closed
1 month ago
5